We derive performance limits for two closely related communication scenarios involving a wireless system with multiple-element transmitter antenna arrays: a point-to-point system with partial side information at the transmitter, and a broadcast system with multiple receivers. In both cases, ideal beamforming is impossible, leading to an inherently lower achievable performance as the quality of the side information degrades or as the number of receivers increases. Expected signal-tonoise ratio (SNR) and mutual information are both considered as performance measures. In the point-to-point case, we determine when the transmission strategy should use some form of beamforming and when it should not. We also show that, when properly chosen, even a small amount of side information can be quite valuable. For the broadcast scenario with an SNR criterion, we find the efficient frontier of operating points and show that even when the number of receivers is larger than the number of antenna array elements, significant performance improvements can be obtained by tailoring the transmission strategy to the realized channel.

Abstract:The propensity score is a common tool for estimating the causal effect of a binary treatment in observational data. In this setting, matching, subclassification, imputation, or inverse probability weighting on the propensity score can reduce the initial covariate bias between the treatment and control groups. With more than two treatment options, however, estimation of causal effects requires additional assumptions and techniques, the implementations of which have varied across disciplines. This paper reviews current methods, and it identifies and contrasts the treatment effects that each one estimates. Additionally, we propose possible matching techniques for use with multiple, nominal categorical treatments, and use simulations to show how such algorithms can yield improved covariate similarity between those in the matched sets, relative the pre-matched cohort. To sum, this manuscript provides a synopsis of how to notate and use causal methods for categorical treatments.

A computerized three-dimensional (3D) neuroanatomy teaching tool was developed for training medical students to identify subcortical structures on a magnetic resonance imaging (MRI) series of the human brain. This program allows the user to transition rapidly between two-dimensional (2D) MRI slices, 3D object composites, and a combined model in which 3D objects are overlaid onto the 2D MRI slices, all while rotating the brain in any direction and advancing through coronal, sagittal, or axial planes. The efficacy of this tool was assessed by comparing scores from an MRI identification quiz and survey in two groups of first-year medical students. The first group was taught using this new 3D teaching tool, and the second group was taught the same content for the same amount of time but with traditional methods, including 2D images of brain MRI slices and 3D models from widely used textbooks and online sources. Students from the experimental group performed marginally better than the control group on overall test score (P = 0.07) and significantly better on test scores extracted from questions involving C-shaped internal brain structures (P < 0.01). Experimental participants also expressed higher confidence in their abilities to visualize the 3D structure of the brain (P = 0.02) after using this tool. Furthermore, when surveyed, 100% of the students in the experimental group recommended this tool for future students. These results suggest that this neuroanatomy teaching tool is an effective way to train medical students to read an MRI of the brain and is particularly effective for teaching C-shaped internal brain structures.

There is a dearth of robust methods to estimate the causal effects of multiple treatments when the outcome is binary. This paper uses two unique sets of simulations to propose and evaluate the use of Bayesian additive regression trees in such settings. First, we compare Bayesian additive regression trees to several approaches that have been proposed for continuous outcomes, including inverse probability of treatment weighting, targeted maximum likelihood estimator, vector matching, and regression adjustment. Results suggest that under conditions of non-linearity and non-additivity of both the treatment assignment and outcome generating mechanisms, Bayesian additive regression trees, targeted maximum likelihood estimator, and inverse probability of treatment weighting using generalized boosted models provide better bias reduction and smaller root mean squared error. Bayesian additive regression trees and targeted maximum likelihood estimator provide more consistent 95% confidence interval coverage and better large-sample convergence property. Second, we supply Bayesian additive regression trees with a strategy to identify a common support region for retaining inferential units and for avoiding extrapolating over areas of the covariate space where common support does not exist. Bayesian additive regression trees retain more inferential units than the generalized propensity score-based strategy, and shows lower bias, compared to targeted maximum likelihood estimator or generalized boosted model, in a variety of scenarios differing by the degree of covariate overlap. A case study examining the effects of three surgical approaches for non-small cell lung cancer demonstrates the methods.

While there is substantial evidence that adults who violate gender stereotypes often face backlash (i.e. social and economic penalties), less is known about the nature of gender stereotypes for young children, and the penalties that children may face for violating them. We conducted three experiments, with over 2000 adults from the US, to better understand the content and consequences of adults’ gender stereotypes for young children. In Experiment 1, we tested which characteristics adults (N = 635) believed to be descriptive (i.e. typical), prescriptive (i.e. required), and proscriptive (i.e. forbidden) for preschool-aged boys and girls. Using the characteristics that were rated in Experiment 1, we then constructed vignettes that were either ‘masculine’ or ‘feminine’, and manipulated whether the vignettes were said to describe a boy or a girl. Experiment 2 (N = 697) revealed that adults rated stereotype-violating children as less likeable than their stereotype-conforming peers, and that this difference was more robust for boys than girls. Experiment 3 (N = 731) was a direct replication of Experiment 2, and revealed converging evidence of backlash against stereotype-violating children. In sum, our results suggest that even young children encounter backlash from adults for stereotype violations, and that these effects may be strongest for boys.